Back to Skills

PDF Reader

Official

by Anthropic

pdfdocument-processingtext-extractionocrform-fillingpythondata-extractionautomation

The PDF skill teaches Claude how to handle any PDF-related task end to end. It covers reading and extracting text with layout preservation, pulling tables into structured data, merging or splitting documents, rotating pages, adding watermarks, password-protecting files, extracting embedded images, and running OCR on scanned PDFs to make them searchable. For PDF form filling it includes a full scripted workflow: detecting whether a form has fillable fields, extracting field metadata, converting pages to images for visual analysis, and writing values back to both fillable and non-fillable forms using coordinate-based text annotations. Internally it uses Python libraries (pypdf, pdfplumber, reportlab, pytesseract) and command-line tools (pdftotext, qpdf, pdftk), with eight ready-to-run Python scripts bundled in the skill.

Installation

/plugin install document-skills@anthropic-agent-skills

Key Features

  • Extract text and tables from PDFs with layout preservation using pdfplumber
  • Merge, split, and rotate PDF pages with pypdf
  • Create new PDFs programmatically with reportlab including multi-page styled output
  • Fill PDF forms — both fillable fields and non-fillable forms via coordinate-based annotations
  • OCR scanned PDFs using pytesseract and pdf2image to make them searchable
  • Encrypt, decrypt, and watermark PDFs with password protection support

Use Cases

  • Extract structured tables from research reports or invoices into Excel or CSV format
  • Automate filling out government or HR forms including non-fillable scanned documents
  • Merge dozens of individual-page PDFs into a single document for archiving
  • Make scanned document archives searchable by running OCR across pages
  • Generate new PDF reports programmatically from data with multi-page styled layouts

Related Resources

Weekly AI Digest